How to write CUDA global function for this? -

i want convert following function cuda.

void fun() {     for(i = 0; < terraingridlength; i++)     {        for(j = 0; j < terraingridwidth; j++)         {              //code of function        }     } } 

i wrote function :

__global__ void fun() {     int = blockidx.x * blockdim.x + threadidx.x;     int j = blockidx.y * blockdim.y + threadidx.y;      if((i < terraingridlength)&&(j<terraingridwidth))     {            //code of function     } } 

i declared both terraingridlength , terraingridwidth constants , assigned value 120 both. , calling function like

fun<<<30,500>>>()

but not getting correct output.

is code wrote correct?.i didn't understood parellel execution of code.please explain me how code work , correct me if done mistakes.

you use y dimension means using 2d array threads, cannot invoke kernel only:

int numblock = 30; int numthreadsperblock = 500; fun<<<numblock,numthreadsperblock>>>() 

the invocation should be: (note blocks have 2d threads)

dim3 dimgrid(grid_size, grid_size); // 2d grids size = grid_size*grid_size dim3 dimblocks(block_size, block_size); //2d blocks size = block_size*block_size   fun<<<dimgrid, dimblocks>>>() 

refer cuda programming guide further info, , if want 2d array or 3d, better use cudamalloc3d or cudamallocpitch

as of code, think work (but haven't tried though, hope can grab idea this):

//main dim3 dimgrid(1, 1); // 2d grids size = 1 dim3 dimblocks(width, height); //2d blocks size = height*width  fun<<<dimgrid, dimblocks>>>(width, height)  //kernel __global__ void fun(int width, int height) {     int = blockidx.x * blockdim.x + threadidx.x;     int j = blockidx.y * blockdim.y + threadidx.y;      if((i < width)&&(j<height))     {            //code of function     } }